Construction Innovation / Manuscript ID CI-Jun-2012-0033 ACCEPTED for publication: 26-Feb-2013

Performance-based procurement construction projects

for

low-disturbance

bridge

Rizal Sebastian1, Christina Claeson-Jonsson2, Roberto Di Giulio3 1

Senior Research Scientist TNO, Expertise Centre of Building and Civil Engineering Van Mourik Broekmanweg 6, 2627AK, Delft, The Netherlands ([email protected])

2

R&D Manager NCC Construction Sverige AB, NCC Engineering Gullbergs Strandgata 2, 40514 Gothenburg, Sweden ([email protected]) Corresponding Author

3

Professor and Head of Department University of Ferrara, Department of Architecture Via Quartieri 8, 44100 Ferrara, Italy ([email protected])

Author biographies: 1

Dr. Rizal Sebastian has over 13 years of research and professional experience in architectural design, Building Information Modelling (BIM), and project management. At present he is a Senior Research Scientist at TNO, the research institute for applied sciences in the Netherlands, and he coordinates the EU FP7 project PROFICIENT. Prior to his current position, he worked as an architect, project manager and consultant at ARCADIS, an international engineering and management firm. Rizal holds a PhD degree in Architectural Design Management from Delft University of Technology in The Netherlands (2007). He graduated with Distinction in MSc in Construction Management at London South Bank University in the UK, in conjunction with HAN University in The Netherlands (2000). He received the Bachelor degree in Architectural Design with Cum Laude from Bandung Institute of Technology in Indonesia (1998).

2

Dr. Christina Claeson-Jonsson has over 15 years’ experience in research activities, design and construction processes. Currently she is the R&D Manager of NCC Engineering at NCC Construction Sweden AB and holds a position as adj. professor at Construction Management, Chalmers University of Technology. Her research mainly focuses on building structures, industrialised construction processes, and energy sustainability. She is the coordinator of EU-funded research projects EBOB (FP5), InPro (FP6), E2ReBuild (FP7), and had a leading role in ManuBuild (FP6). Christina holds a PhD degree (1998) and MSc degree (1991) in Structural Engineering from Chalmers University of Technology, Sweden.

3

Prof. Roberto Di Giulio has over 25 years’ experience in architectural design and research activities. Currently he is full professor of Architectural Technologies at the Faculty of Architecture and Head of the Department of Architecture at the University of Ferrara (Italy). He is a Partner at Ipostudio Architects, a planning and research organisation operating in the fields of architecture, design and research, in Florence since 1982. His research covers a broad range of studies from the industrialization of building process and building design methodologies to the investigation on real estate maintenance and management. Roberto holds a PhD degree in Architectural Technologies from the University of Rome La Sapienza (1989) and MSc degree in Architecture from Faculty of Architecture at the University of Florence (1982).

Acknowledgement: This paper refers to the currently ongoing collaborative research project titled PANTURA, co-funded by the European Commission within the EU FP7 programme (www.pantura-project.eu). The authors are lead researchers within the research consortium. The case study data are provided by NCC, Swedish Transport Administration, and AICE Consulting Srl.

1

Abstract: Purpose: This paper aims to introduce a method of performance-based procurement, based on the Most Economically Advantageous Tender (MEAT), for low-disturbance bridge construction projects in urban environment. Design/Methodology/Approach: The first part of this paper reviews the key performance indicators (KPIs) of low-disturbance construction and the procurement procedure based on the MEAT principles. The second part reflects on two actual bridge projects (the Rotebro bridge in Sweden and the Arno River bridge in Italy) as observatory case studies to analyse how clients and contractors can implement the KPIs in MEAT. Findings: The research findings demonstrate the possible inclusion of the KPIs of low-disturbance construction into the MEAT criteria. The MEAT principles can then be used in combination with either a traditional or an integrated procurement strategy. Research limitations/implications: The implementation of MEAT to achieve low-disturbance construction projects is considerably new and still requires an empirical validation. A further elaboration of the procurement strategy within the EU regulatory framework is strongly recommended in order to assure the broader impacts of sustainable construction. Practical implications: The findings and recommendations support the practical development and the use of MEAT in construction projects in the EU. Originality/Value: This paper presents on-going investigation within the FP7 collaborative research project “PANTURA”, which addresses the actual research agenda of the European Commission on low-disturbance and urban-friendly civil infrastructure projects.

Keywords: Sustainable construction, low disturbance, urban friendly, bridge projects, Most Economically Advantageous Tender (MEAT), public procurement

Article classification: Research paper

2

Introduction Bridges form the backbone of an urban civil infrastructure system. A large number of bridges are located in densely populated urban areas and they vitally contribute to the transport and mobility performance in many cities. Bridges also have a very high asset value as key objects and landmarks of the city architecture. Along with the demand for higher loads and more intensive utilisation of bridges, the need to construct new bridges and to strengthen, repair and upgrade the existing ones has been increasing significantly. Unfortunately, the construction and maintenance activities of bridges in the cities often become a major cause of accessibility and safety problems, which include: nuisance, noise and vibration; dust and air pollution; disturbances and disruptions; traffic jams, delays and accidents. Another main challenge for urban bridge (re)construction projects is the inefficient use of resources, i.e. materials, energy, equipment and labour. Josephsson and Saukkoriipi (2005) found the most common errors occurred in design (26%), site planning and management (25%), production (20%), material supply (17%), client’s decision (6%), inefficient equipment use (3%), and other aspects (3%). Therefore, the urban communities and governments urgently call the construction industry to adopt an innovative strategy for low-disturbance construction processes in large-scale infrastructure projects. This paper aims to introduce a procurement strategy for low-disturbance construction projects, which is essential for defining the common project goals; distributing the responsibilities and risks; selecting and commissioning the most appropriate contractor; and implementing the most sustainable solutions and construction techniques. Since most urban infrastructure projects belong to public clients, the proposed strategy needs to meet public procurement regulations. In the EU, as described in the Public Procurement Directives, contracting authority can choose to apply the Lowest Price tender or the Most Economically Advantageous Tender (MEAT) (European Parliament, 2004; Martineau, 2010; SIGMA, 2011; JISC Procureweb, 2013). As opposed to the Lowest Price tender, MEAT enables the contracting authority to take account of criteria that reflect qualitative, technical and sustainable aspects of the tender submission as well as price when reaching an award decision. MEAT allows contracting authorities to take into account innovation or innovative solutions, and it is typically used when quality is important for the contracting authority. A comprehensive empirical study over 180 procurement processes in Finland, Sweden, and Denmark where MEAT was used has been conducted by Parikka-Alhola et al. (2006). It is evident that the MEAT has been widely used in the EU and acknowledged within the current regulatory framework, especially for Green Purchasing Projects. This fact encourages the attempt to establish a broader implementation of the MEAT for bridge reconstruction projects. MEAT becomes particularly relevant when low-disturbance construction is prioritised in the public procurement of bridge construction projects (Sebastian, 2011). Up to now, there are challenges to MEAT implementation as described by Lavér et al (2011), being that current regulations on public and utilities procurements, reflecting the essential parts the EU directives, do not stipulate how the MEAT evaluation procedure should be conducted. In accordance with the fundamental principles of public and utilities procurement, a contracting authority or entity must, in a transparent and predictable manner, present the method for the evaluation procedure. However, the identification of MEAT might be a difficult task, especially in the context of low-disturbance construction projects: What criteria should be used? How should the evaluation process be performed? How should an evaluation model be constructed and applied in order to comply with the fundamental principles of transparency, predictability and equal treatment? This paper focuses on research towards the implementation of MEAT, especially in response to the current knowledge gap. It reports the literature and empirical studies within the on-going EU FP7 collaborative research project titled “PANTURA” (www.pantura-project.eu). Following this introduction section, the second section of this paper covers theoretical review on two subjects: 1) the key performance indicators (KPIs) of low-disturbance urban construction; and 2) the implementation concept of MEAT, especially regarding the use of low-disturbance KPI as MEAT criterion as well as the methods for weighting and evaluation. The third section of this paper analyses two real cases of bridge construction projects with lowdisturbance objectives, i.e. the Rotebro bridge in Sweden and the Arno River bridge in Italy. In the final section of this paper, discussions, conclusions and recommendations are provided.

3

Theoretical review Key performance indicators (KPIs) for low-disturbance construction projects KPIs for low-disturbance urban projects are crucial for the stakeholders, especially the (public) client and the contractor, for deciding a project strategy based on reliable data throughout the project’s life-cycle. The KPIs can be either used as predictive indicators or retrospective indicators. The predictive indicators are based on the existing models and estimations to be performed before the start of the construction. Using the predictive indicators, the project’s targets and goals can be defined. The retrospective indicators are computed at the completion of each stage of the project. Based on these indicators, the actual performance of the project can be validated against the previously defined goals. These indicators are also useful for the impact assessment of the project and for the benchmarking over a number of similar projects (PANTURA, 2011). There is a range of existing methods and tools to measure sustainability in construction, each of them comes with a set of indicators. During the development of KPIs for low-disturbance construction projects, these instruments were critically reviewed. Greenroads (Muench et al, 2011), CEEQUAL (CIRIA and Crane Environmental, 2008), ISO CD 21929-2 (2010), Ugwu et al (2006) and LEED (United States Green Building Council, 2011) were found to be the most relevant for use in civil infrastructure projects like bridge construction, and with regards to sustainability at the construction site (PANTURA, 2011). However, there is a knowledge gap regarding the subject of disturbance to the urban environment during the construction period. The indicator selection and evaluation process was performed in accordance with a framework derived from the UN framework for the development of sustainability indicators, United Nations (2007). This process involved the initial determination of relevant indicators followed by an analysis of data availability. To determine relevance, an extensive literature review was conducted of existing indicator sets and sustainability rating systems used for similar projects. The literature review found that no existing set of indicators adequately addresses the disturbance aspect during construction. It also revealed that if too many indicators are used, the decision-making process may become too complex and inefficient, PANTURA (2011). Ultimately, sixteen internationally recognized indicators were selected and a profile developed for each of these. The candidate indicators were sent for review to a panel of international experts and stakeholders. The results of the survey were analysed in a transparent and consistent fashion and the list was finally narrowed down to 11 specific PANTURA KPIs with respect to disturbance during the construction stage, PANTURA (2011). These were then grouped into four themes: 1) mobility (traffic flow); 2) safety (residents, construction workers); 3) accessibility (duration of construction on site); and 4) environment (noise, dust, waste, reused or recycled materials, total use of materials, life cycle costs, and greenhouse gas emissions). The KPIs associated with the four themes were designed to become the objective references to be used in a performance-based procurement method for low-disturbance construction projects.

4

Theoretical concept of MEAT implementation: the use of low-disturbance KPIs In the last few years, public governments in many countries have started using innovative supplier selection methodologies that are drawn from the established practices of private sector (de Boer et al., 2001; Panayiotou et al., 2004; Erridge and Callender, 2005; Love et al., 2011). The supplier selection is a multiobjective problem conclude Weber et al. (1991), after having reviewed 74 article, since the selection process includes conflicting objectives, such as quality, quantity, delivery, performance, capacity, communication, service, geographical location, in addition to lowest price (Degraeve et al., 2000; Morlacchi, 1999; Falagario et al, 2012). However, differently from the private sector, the awarding committee of the client must follow prescribed procedures and maintain transparency in public procurement (Panayiotou et al., 2004). The contracting authority can chose to implement MEAT in either traditional procurement or integrated procurement. The traditional procurement is widely known by the sequential ‘design–bid–build’ process. It is characterised by a contractual separation between design and construction responsibilities. Based on the design and technical specifications, the client organises a public tender to appoint a contractor. The integrated procurement is commonly identified as the ‘design and build’ process, which can include the extended forms of design–build–maintain (DBM), design–build–finance–maintain (DBFM), and design– build–finance–maintain–operate (DBFMO). In the integrated procurement, the client establishes a contract with a single party (a main contractor or a consortium of designer and contractor), which assumes the full responsibility for both designing and constructing the project. The tender procedure, prior to contracting, is based on the functional programme of requirements, instead of the detailed design and technical specifications (Joint Contract Tribunal, 2009; Sebastian et al., 2007; Trans-IND, 2011). When traditional procurement is used, the KPIs for low-disturbance urban projects are mainly applied as predictive indicators associated with the design requirements. Based on norms, references, and simulations, the lead designer presents the most optimal design solution that is expected to achieve the required performance. The appointed contractor subsequently carries out the construction. The client’s task is to assure that the contractor follows the design and its specifications accurately. If the targeted performance measured against the KPIs of low-disturbance urban projects is not achieved, discussions will arise whether the cause of this sub-optimal performance lies in the design solution or in the construction. Either the designer or the contractor can be held accountable. The extent of their liabilities depends on the applicable legal framework, but usually it is limited to the maximum value of their commissions. In most cases, the client also takes the cost and the risks to restore the project. When integrated procurement is used, the KPIs for low-disturbance urban projects are applied both as predictive as well as retrospective indicators. As predictive indicators, the KPIs are used in correspondence to the functional requirements prior to the design-and-build tender. The contractor proposes the most optimal design and construction solutions. After the design–and–build contract is awarded, the contractor is fully responsible to manage the design and construction processes, and it has the freedom to deploy its preferred methods. As retrospective indicators, the KPIs are used at the agreed milestones for monitoring the performance of the intermediate results during the project duration. The payment to the contractor can be approved based on the achievement of the targeted performance. If the targeted performance is not achieved, the contractor must restore the design and construction at its own cost. Special attention should be given to how the contracting authority has to select the best MEAT criteria to ensure the low-disturbance performance. The EU Public Procurement Directives do not set out an exhaustive list of the sub-criteria which may be chosen, thus the contracting authority can chose other ones besides (or instead) price, quality, delivery date, product life, aesthetic, after sales service, etc. The KPIs for lowdisturbance construction projects can, therefore, be transformed into appropriate MEAT criteria provided that they fulfil the requirements stipulated in Directives (European Parliament, 2004). In practice, the criteria that a contracting authority may apply to determine the MEAT must be chosen in such a way that they match the contract specifications. All specifications subject to evaluation should have criteria associated with them (SIGMA, 2011).

5

Theoretical concept of MEAT implementation: the weighting factors and evaluation method ‘The best value for money’ is the goal of any economic transaction. This also applies for the project delivery strategy in construction, particularly during the procurement process that includes a tender procedure. Many authors have argued about the need for differentiating cost and benefit criteria in the evaluation of tenders, Topcu (2004), Falagario et al. (2012). Based on this principle, the MEAT evaluation methods were elaborated to clarify the winning chance of the bid with the most optimal value–price ratio (Dreschler, 2009). The MEAT needs an objective tender evaluation method to comprehensively evaluate and compare the tenders, instead of evaluating and selecting the winning tender based solely on the lowest price. Other aspects (e.g. innovation and sustainability) that can add values to the project are also taken into account either directly or indirectly. By its aim for value–price optimisation, the MEAT differs from the tender that focuses only on price minimisation (i.e. lowest price bid for fixed-requirements) or the tender that focuses only on value maximisation (i.e. fixed-price design contest). The MEAT can be supported by analytical evaluation methods as described by Dreschler (2009), Falagario et al, (2012), e.g. point system, ratio system, and price correction system. Table 1 gives an illustration of different evaluation methods. When the ‘point system’ is used, all aspects of the bid –including the offered price and the added values beyond the minimum tender requirements– are translated into points according to an objective calculation reference. The bid with the highest point becomes the winner. When the ‘ratio system’ is used, the monetary basic value of the minimum tender requirements is determined first. Then, the added value of each bid above the minimum tender requirements is determined, and added to the basic value (i.e. total value = basic value + added value). The bid with the highest value/price ratio becomes the winner. When the ‘price correction system’ is used, the added value of each contractor’s bid above the minimum tender requirements is determined. The offered price will then be adjusted depending on its added value (i.e. corrected price = offered price – added value for the project). The bid with the lowest corrected price becomes the winner. Since low-disturbance bridge construction projects differ from purchasing projects in certain aspects, this paper proposes the following steps for the implementation of MEAT: - First, MEAT components of low-disturbance aspects should be defined. These components can refer to the low-disturbance KPIs from the PANTURA research project as discussed in the previous section of this paper. - Second, the weighting factors to each MEAT component should be assigned. Although standard MEAT components are applied, the weighting factors may be different in each project depending on the project type, urgency, stakeholder’s objective, etc. - Third, a cohesive calculation reference should be introduce to translate the MEAT components into points (i.e. when the ‘point system’ is used) or into monetary values (i.e. added value when the ‘ratio system’ is used or real cost saving when the ‘price correction system is used). - Fourth, the most appropriate MEAT evaluation method should be decided depending on the project delivery strategy. ‘Point system’ or ‘ratio system’ is usually preferable for integrated project delivery strategy, and ‘price correction system’ is suitable for lowest-price procurement within the traditional strategy. - Fifth, the MEAT evaluation and contract award procedures should be clarified and adequately explained to all actors involved in the procurement process through written and/or oral information sessions and competitive dialogs, if relevant.

6

MEAT analysis mechanism and calculation reference for tender evaluation purposes Point system The lowest price bid gets 100 basic points. The cost saving impact is considered as an added value to be added on top of the basic points. The bid with the most points wins.

Fictional bid A Offered Cost saving price: impact or added value: 100 million 0 million

Fictional bid B Offered Cost saving price: impact or added value: 110 million 15 million

Fictional bid C Offered Cost saving price: impact or added value: 120 million 20 million

Basic points = 100 points [the lowest price gets 100 points]

Basic points = 90 points [the offered price is 10 million more expensive than the lowest price in Bid A]

Basic points = 80 points [the offered price is 20 million more expensive than the lowest price in Bid A]

Added value = 0 points

Added value = 15 points

Added value = 20 points

Total points = 100 points

Total points = 90 + 15 Total points = 80 + 20 = 105 points = 100 points Bid B earns the most points, and thus it becomes the winner

Ratio system The basic value of the minimum tender requirements is 100 million. The cost saving impact of each bid is considered as an added value. The bid with the highest ratio of total value divided by the offered price wins. Price correction system The real cost saving impact of each bid is deducted from the offered price. The bid implies the lowest cost for the client wins.

Total value = 100 + 0 = 100 million

Total value = 100 + 15 = 115 million

Total value = 100 + 20 = 120 million

Value/price ratio = 100/100 = 1.00

Value/price ratio = 115/110 = 1.05

Value/price ratio = 120/120 = 1.00

Bid B has the highest price/value ratio, and thus it becomes the winner

Real total cost for the client = 100 + 0 = 100 million

Real total cost for the client = 110 – 15 = 95 million

Real total cost for the client = 120 – 20 = 100 million

Bid B implies the lowest cost for the client, and thus it becomes the winner

Table 1: Illustrative examples of the MEAT evaluation methods

7

Case studies Case overview In this section, the intended use of the MEAT approach as analysed based on theoretical concepts are reflected on two real cases of bridge construction. The case overview is shown in Table 2. The selected cases are derived from up-to-date practice of the consortium partners in the EU research project “PANTURA”. Low-disturbance is a part of the project goals, yet this aspect has not been fully accommodated in the actual tender procedure. There is much interest from the stakeholders of both projects to improve the tender procedure of similar projects in the future. By analysing the two cases, this paper highlights the advantages and disadvantages of MEAT, and recommends the implementation of low-disturbance KPIs in MEAT.

- Location - Scope of project - Client - Contractor - Cost and duration

- Project delivery strategy

Case 1: E4 bridges at Rotebro Sollentuna, Sweden Demolition and replacement of two existing bridges across the railway Swedish Transport Administration NCC Construction Sweden Cost approximately 325 million SEK (40 million Euros); duration 3 year. The construction commenced in February 2011 and is expected to be completed in spring 2014. Design-and-build with public tender. Primary tender criterion: Lowest price with a bonus scheme if the work is completed ahead of the original deadline.

Case 2: Bridge over the Arno River at Empoli Empoli (Florence), Italy Demolition and replacement of an existing bridge over the Arno river Provincial Authority of Florence Spinosa Costruzioni Generali S.p.A. Cost approximately 7,9 million Euros; duration required in the tender 3 years. 50% of the bridge has been completed and opened to traffic in July 2011, the remaining work is in progress. Design-and-build with public tender. Primary tender criterion: Economically Most Advantageous Tender based on criteria descripted in the tender specifications.

Table 2: Overview of the two observatory case studies Case 1: E4 bridges at Rotebro in Sollentuna, Sweden More than 70 thousands motorists on the E4 Highway in Sweden cross the two bridges (Eastern and Western bridges) at Rotebro each day in both directions (Figure 1). Underneath the bridges, 600 trains per day pass the bridges. These existing bridges are to be demolished and replaced by new ones. The biggest challenge is to carry out the demolition and construction work while allowing the highway and railway traffic to pass the site with as few disruptions as possible. Since the site partially lies within the primary zone of the water protection area adjacent to the Stockholm Ridge, another major challenge is to assure the groundwater quality and to fulfill the requirement for environmental sampling throughout the construction project. A CEEQUEL environmental classification is to be applied for the first time in Sweden on this bridge project.

Figure 1: Existing E4 bridges at Rotebro in Sollentuna, Sweden (source: NCC)

8

Due to the complexity of the project, the client –the Swedish Transport Administration– used the design-andbuild project delivery strategy. It aimed to allow more freedom and to allocate a greater responsibility to the contractor. The tender invitation and description encouraged the tenderers to propose innovative solutions and allowed them to submit alternative tenders. The tenders were to be evaluated primarily based on the prices, but a bonus would be given to the contractor if the construction period could be shorter than the foreseen duration by the client. The winning tender excelled in terms of lowest price as well as low-disturbance urban construction. The contractor proposed a method through which the new Eastern bridge would be constructed in an adjacent location. This bridge would be used as a temporary detour during the construction period. In this manner, there would be two separate and operational bridges throughout the majority of the construction period, each with three lanes opened for traffic. In term of low-disturbance, this solution meant: - Better solution for traffic flow exceeding the client requirement. Highest mobility capacity during construction. The winning contractor’s proposal would allow a higher maximum speed of 90 km/h, higher than 70 km/h as required by the client; over six lanes, more than five lanes as requested by the client –except during 3 weeks in the summer period when only two lanes would be opened in each direction with a maximum speed of 70 km/h. - Better accessibility. The contractor planned to finish the construction 3 weeks ahead of the original deadline given by the client.

Case 2: Bridge across the Arno River at Empoli (Florence), Italy This project regards the replacement (through demolition and construction) of the bridge over the Arno river that links the villages of Empoli and Sovigliana on the Provincial Highway 13 in the vicinity of Florence. The bridge was built in the early 1950s, and it underwent a large repair after the damage caused by the flood in 1966. Recent investigations and tests have reported critical structural weaknesses. Therefore, the client – the Provincial Authority of Florence– defined the following requirements: - to build a bridge suitable for safe crossing of Arno river complying with current regulations on construction, seismic and water safety; - to increase the capacity of traffic flow, and to provide a new bridge with a bicycle path; - to carry out the construction works while allowing the crossing of the river by vehicles and pedestrians with as little disturbance as possible. Once the functional programme of requirements was established, the client decided to define the strategy and find the most optimal technical solutions regarding cost, construction duration, and long-term performance. The procurement process was divided in two stages: 1) procurement of design and engineering; and 2) tender for the construction work. As the design and engineering contract was below the stipulated threshold of the Italian legislation (EUR 100,000), the selection was done through a negotiating procedure. The selected firms presented two possible solutions in the preliminary design. The design illustration is shown in Figure 2. Five candidate firms, selected from a list filed at the Province of Florence, were invited to join the first tender procedure to generate the preliminary design and the technical design (definitive design) respectively. The chosen solution, which was also the lowest-price offer, showed the best cost/benefit ratio.

9

Figure 2: View of the new bridge across the Arno River under construction (source AICE) The chosen solution was further detailed in the definitive design and used together with the technical specifications for the tender. After the approval of the definitive design, the Provincial Authority of Florence started the second tender procedure to construct the bridge. The tenderers’ proposals should focus on price, delivery terms, quality of technical solutions, and procedures related to the organization of the construction site, quality and performance of materials and finishing. The tender was evaluated based on MEAT criteria, as shown in Table 3. A special commission was established to evaluate and compare the contractors’ offers. Economic offer E1: Percentage of cost saving (CS) compared to the tender price. CSmax = maximum Cost Saving offered. CSc = Cost Saving offered by the contractor to be scored. Technical offer T1: Delivery terms; reduction of time compared to the time scheduled in the project plan developed by the client. Dr = number of days less than scheduled by the client. Time scheduled in the client’s project plan = 1100 days. Maximum time reduction allowed = 200 days. T2: Quality of works; improvements of technical solutions focused on low impact of the construction site. T3: Design quality; improvement of quality and performances of materials and finishing Maximum total score Formula to calculate the score: E1 score = 50 x (1 – CSmax / 1 – CSc) T1 score = 20 x Dr/200 T2 and T3 scores were calculated based on the relative comparison between the received tenders.

Maximum score 50

20

20 10 100

Table 3: MEAT criteria in the bridge project across the Arno River at Empoli (Florence), Italy

10

Cross-case analysis The two cases exemplify the intention for low-disturbance bridge construction based on traditional procurement (design-bid-build), i.e. case of the bridge project in Italy, and integrated procurement (design and build), i.e. case of the bridge in Sweden. In the Arno River bridge case, two tender procedures were carried out: the first tender was for design effort, and the second one for construction work. MEAT criteria were used in the tender for construction work. The main advantages and disadvantages of MEAT implementation in traditional procurement can be analysed as follows. The advantages include: more certainty about the technical result as the client was able to assess the possible solutions and determine the strategy since the preliminary design stage. This was considered very important in this bridge project because the architectural quality of the bridge was a crucial factor in the eyes of the client. The other advantages are: client’s control and possible intervention regarding the costs and project duration (KPI accessibility); and the possible deployment of the available tools and client’s competence for checking and selecting the technical proposals. The disadvantages are: the cost incurred at the client to have the definitive design and some alternatives produced before the tender; and the limited possibility for the appointed contractor to propose a different design approach – including a more innovative one – at the tender. The choice of the client to develop a detailed design before the tender was due to its obligation to define and verify the economic, technical and environmental impacts of the design solution at the early phase. Further advantages could be achieved focusing on the client’s needs and providing a more detailed description of the quality parameters to be evaluated in the tender. In the Rotebro bridge case, MEAT was not used as the tender was based on lowest-price bidding principles. Nevertheless, the winning offer was proven to have the lowest price, as well as the most advantages in terms of low-disturbance solutions. The contractor’s offer proposed a better traffic flow (KPIs mobility and safety) during construction leading to less exhaust gas emission and noise (KPI environment). Furthermore, construction is planned to finish three weeks ahead of the original deadline (KPI accessibility). The main drawback is the long-term lifecycle aspects were not included in the tender evaluation, partly due to the inability to do so without using MEAT criteria. Thus, more research is needed on how the MEAT using lowdisturbance KPIs can increase the possibilities for generating offer that bring added values in respect of minimizing disturbance during construction as well as maximizing benefits during the bridge’s lifecycle.

11

Discussions Based on the case study analysis, this paper proposes five steps to implement MEAT for low-disturbance bridge construction projects. The first step is establishing the MEAT protocols by deciding which indicators from the set of generic KPIs for low-disturbance construction should be included as MEAT criteria. Taking the Rotebro bridge case as example, mobility and duration of construction are the most important aspects regarding disturbance. Other aspects include waste and dust emission, which may be considered in relation to maintaining the environmental quality, especially the groundwater. Furthermore, the cost to design and construct the project is the primary tender criterion, and can be considered as a part of the lifecycle cost of the new bridges. Taking the Arno River bridge as example, there were several parameters potentially related to disturbance besides the duration of construction (20% of the total score) and price (50% of the total score), which are: quality of technical solutions and organisation of construction site (20%) and quality of design (10%).). The second step is to assign the priorities and the associated weighing factors to the selected MEAT components. Reflecting on the Rotebro bridge project, cost (tender price) would get the most dominant weighing factor, followed by accessibility (in relation to the bonus scheme), mobility (traffic flow), and environment. For example, the weighing factors for these MEAT components are: 80, 10, 5, and 5 (the total is 100). Reflecting on the Arno River bridge project, the application of the low-disturbance KPIs could better detail the parameters regarding disturbance giving them a prioritization related to the purposes and expectations of the local authority. In the existing practice, unlike the detailed calculation of the score given for the delivery time reduction, it was not sufficiently detailed how the scores for parameters T2 (improvements of technical solutions focused on low impact of the construction site) and T3 (improvement of quality and performances of materials and finishing) should be calculated. The third and fourth steps are to translate the value of the selected MEAT components into points or monetary amount, and then to select the MEAT evaluation method. Since the tender price is the leading aspect, and the bonus is not to be calculated in the tender price, the ratio system could be most appropriate in case MEAT is implemented in the Rotebro bridge project. Using the ratio system, the added values of better mobility, shorter duration and less environmental effects should be monetized. The value-price ratio of each bid is then determined by the total value (tender price plus added values) divided by the tender price; therefore, the lowest tender price is still the most determining factor to achieve the highest value-price ratio. In the tender procedure of the Arno bridge project, the application of low-disturbance KPIs was defined by the local authority, and the weighting factors were assigned for the different criteria. The fifth step is to formally describe the MEAT protocols and explain them to all tenderers. Whenever necessary and allowed within the applicable legal framework, competitive dialogs can be organised as the opportunity for open discussions between the client and all tenderers. Such dialogs are very important to clarify the protocols and the tender evaluation process, especially when the most parties experience the MEAT for the first time.

12

Conclusions Low-disturbance construction is an essential aspect of a sustainable project, and has become significantly important in bridge projects in a high density urban environment. Until now, it is difficult to define, quantify and agree upon the performance requirements of low-disturbance construction. This paper investigates the usefulness to introduce a set of objective KPIs for this purpose in the MEAT phase. It is shown that the MEAT can be implemented in combination with either the traditional (design–bid–build) or integrated (design-and-build) procurement. Special attention should be given when setting-up the MEAT protocols, particularly regarding the way MEAT criteria are defined, the decision on the priorities and the associated weighing factors, and the deployment of the most appropriate evaluation method. Over-simplifying or overcomplicating the MEAT protocols can have a negative effect for the project delivery strategy, and can lead to a disputable tender result that compromises the performance achievement. Two case studies were used to highlight the advantages and disadvantages with the proposed MEAT protocol. In order to create a wider impact at the EU or international level, the standardisation of the KPIs for lowdisturbance construction projects in combination with MEAT is necessary. Standard guidelines are important to provide (public) clients with solid references and valid justifications when determining the weighting factors for their projects. Further research should clarify which elements of the KPIs and the MEAT should be made generic, and which others depend on the specific project context, taking into account the similarities and differences of procurement frameworks in the EU and in different countries.

13

References Boer, L. de, Linthorst, M., Schotanus, F., Telgen, J. (2006), “An analysis of some mistakes, miracles and myths in supplier selection”, in Proceedings of the 15th IPSERA Conference, San Diego, USA. CIRIA and Crane Environmental (2008), CEEQUAL Assessment Manual for Projects Version 4, CIRIA and Crane Environmental, London, UK. Degraeve, Z., Labro, E., Roodhooft, F., (2000), “An evaluation of vendor selection models from a total cost of ownership perspective”, European Journal of Operational Research 125, pp. 34–58. Dreschler, M. (2009), Fair Competition: How to Apply the Economically Most Advantageous Tender (MEAT) Award Mechanism in the Dutch Construction Industry, PhD Dissertation, Delft University of Technology, Delft, The Netherlands. Erridge, A., Callender, G., (2005), “Introduction to the special issue on public procurement”, Journal of Purchasing and Supply Management, 11 (5–6), pp. 209–211. European Parliament (2004), Directive 2004/18/EC of the European Parliament and of the Council of 31 March 2004 on the Coordination of Procedures for the Award of Public Works Contracts, Public Supply Contracts and Public Service Contracts, European Parliament, Brussels, Belgium. Falagario, M., Sciancalepore F., Costantino N., Pietroforte, R., (2012), “Using a DEA-cross efficiency approach in public procurement tenders”, European Journal of Operational Research, 218, pp. 523–529 ISO CD 21929-2 (2010), Sustainability in Building and Civil Engineering Works – Sustainability Indicators Part 2: Framework for the Development of Indicators for Civil Engineering Works, ISO, Geneva, Switzerland. JISC Procureweb (2013), “Most economically advantageous tender (MEAT)”, available at: http://help.procureweb.ac.uk/goodpracticeguide/1302legislation/1101eu_public_proc_legislation/1217award _criteria/1219meat/index.html (accessed 17 January 2013). Joint Contract Tribunal (2009), Deciding on the Appropriate JCT Contract, Sweet and Maxwell, London, UK. Josephson, P.-E., Saukkoriipi, L., (2005), ”Slöseri i byggprojekt – behov av förändrat synsätt”, FoU-Väst Rapport 0507, Sveriges Byggindustrier, Sweden [in Swedish]. Lavér, J., Larsberger, O., (2011), "The Art of Identifying 'The Most Economically Advantageous Tender' The Use of Relative Evaluation Models in Public and Utilities Procurement", available at: http://www.whoswholegal.com/news/features/article/29137/the-art-identifying-the-economicallyadvantageous-tender-use-relative-evaluation-models-public-utilities-procurement/ (accessed 18 April 2012). Love, P.E.D., Edwards, D.J., Irani, Z., Sharif, A., (2012), “A Participatory Action Research Approach to Public Sector Procurement Selection”, Journal of Construction Engineering and Management, 138(3), pp. 311-322. Morlacchi, P., (1999): “Vendor evaluation and selection: The design process and a fuzzy-hierarchical model”, in Proceedings of the 8th International Annual IPSERA Conference, Belfast/Dublin, Ireland. Muench, S.T., Anderson, J.L., Hatfield, J.P., Koester, J.R., Söderlund, M. (2011), Greenroads Manual v1.5. (J.L. Anderson, C.D. Weiland, and S.T. Muench, Eds.). Seattle, WA: University of Washington, available at: http://www.greenroads.org/347/the-rating-system.html (accessed 17 January 2013). Panayiotou, N.A., Gayialis, S.P., Tatsiopoulos, I.P., (2004), “An e-procurement system for governmental purchasing”, International Journal of Production Economics, 90, pp. 79–102. PANTURA (2011), Comparative Analysis of Best Practices: Review of Existing Methods and Development of an Indicator Suite for PANTURA Goals, Project Report D6.4, EU FP7 Project PANTURA. Parikka-Alhola, K., Nissinen, A., Ekroos, A. (2006), “Green Award Criteria in the Most Economically Advantageous Tender in Public Purchasing”, in Proceedings of the 2nd International Public Procurement Conference (IPPC), Rome, Italy, 21-23 September 2006, PrAcademics Press, Boca Raton, FL, USA.

14

Sebastian, R. (2011), “Low Disturbance Urban Projects through Smart Manufacturing and BIM”, in Bartolo, H. (Ed.), Proceedings of International Conference on Sustainable Intelligent Manufacturing (SIM), Leiria, Portugal, 29 June – 1 July 2011, IST Press, Lisbon, Portugal, pp. 37-44. Sebastian, R., Gelderen, K.A. van (2007), “Developing a Model to Support Client’s Decision Making Process on Integrated Contracts”, in Wamelink, J.W.F., Ridder, H.A.J. de (Eds.), Proceedings of 2nd World of Construction Project Management Conference, in Delft, The Netherlands, 22-25 May 2007. SIGMA (2011), Public Procurement: Setting the Award Criteria, SIGMA (joint initiative of the OECD and the European Commission), Paris, France. Topcu, V.I., (2004), “A decision model proposal for construction contractor selection in Turkey”, Building and Environment 39, pp. 469–481. Trans-IND (2011), Current Business Models for Transport Infrastructure Projects, Project Report D9.2, EU FP7 Trans-IND. Ugwu, O.O., Kumaraswamy, M.M., Wong, A., Ng, S.T. (2006), “Sustainability appraisal in infrastructure projects (SUSAIP) Part 1. Development of indicators and computational methods”. Automation in Construction 15, pp. 239 – 251 United States Green Building Council (2011), “Rating Systems in LEED”, http://www.usgbc.org/DisplayPage.aspx?CategoryID=19 (accessed 4 November 2011).

available

at:

United Nations (2007), Indicators of Sustainable Development: Guidelines and methodologies – Third Edition. United Nations publication, ISBN 978-92-1-104577 -2. United Nations, New York. Weber, C., Current, J.R., Benton, W.C., (1991), “Vendor selection criteria and methods”, European Journal of Operational Research 50, pp. 2–18.

15